Sapphire

Article pubs.acs.org/crystal

Epitaxy of Boron Phosphide on Aluminum Nitride(0001)/Sapphire Substrate Balabalaji Padavala,† C. D. Frye,† Xuejing Wang,‡ Zihao Ding,‡ Ruifen Chen,‡ Michael Dudley,‡ Balaji Raghothamachar,‡ Peng Lu,§ B. N. Flanders,∥ and J. H. Edgar*,† †

Department of Chemical Engineering, Kansas State University, Manhattan, Kansas 66506, United States Department of Materials Science and Engineering, Stony Brook University, Stony Brook, New York 11794, United States § Nitride Solutions Inc., 3333 West Pawnee Street, Wichita, Kansas 67213, United States ∥ Department of Physics, Kansas State University, Manhattan, Kansas 66506, United States ‡

S Supporting Information *

ABSTRACT: The boron phosphide (BP) semiconductor has many remarkable features, including high thermal neutron capture cross section of the 10B isotope, making it attractive for neutron detection applications. Effective and efficient neutron detection require BP to also have high crystal quality with optimum electrical properties. Here, we present the heteroepitaxial growth of high quality BP films on a superior aluminum nitride(0001)/sapphire substrate by chemical vapor deposition. The effect of process variables on crystalline and morphological properties of BP was examined in detail. BP deposited at high temperatures and high reactant flow rate ratios produced films with increased grain size and improved crystalline orientation. Narrower full width at half-maximum values of BP Raman peaks (6.1 cm−1) and ω rocking curves (352 arcsec) compared to values in the literature confirm the high crystalline quality of produced films. The films were n-type with the highest electron mobility of 37.8 cm2/V·s and lowest carrier concentration of 3.15 × 1018 cm−3. Rotational twinning in BP due to degenerate epitaxy caused by 3-fold BP(111) on 6-fold AlN(0001) was confirmed by synchrotron white beam X-ray topography. This preliminary study showed that AlN is an excellent substrate for growing high quality BP epitaxial films with promising potential for further enhancement of BP properties.

1. INTRODUCTION Cubic boron phosphide (BP) is a III−V semiconductor with exceptional features such as high bulk modulus (155.7 GPa),1 chemical inertness,2 high temperature stability, high mechanical hardness,3 high thermal conductivity (4.0 W/cm·K),4 and low ionicity.5,6 BP has an indirect wide band gap of 2.0 eV,7−9 and both n-type and p-type conductivities have been reported with high electron and hole mobilities.10,11 10BP is an attractive semiconductor for neutron detectors due to the high thermal neutron capture cross section of the 10B isotope (3840 barns).12 Because BP is composed of elements with low atomic numbers, it enables better discrimination between gamma and neutron radiation. Therefore, the ability to produce both n-type and ptype conductivities by adding dopants and ease of device fabrication makes BP an ideal material for neutron capture, charge creation, and charge collection for a solid state neutron detector. For effective and efficient neutron detection, BP must also have high crystalline quality with optimum electrical properties. Although BP epitaxy on substrates such as silicon,11,13−17 sapphire,18,19 SiC,9,20−22 and GaN/sapphire23 has been studied for more than 45 years, with few exceptions, most previous studies produced BP with poor electrical properties due to the unintentional impurities and structural defects introduced in the films during epitaxy. The primary reasons for generation of these impurities and defects are poor chemical stability of the © XXXX American Chemical Society

substrate, large mismatches of lattice constants and thermal expansion coefficients between the substrate and BP, and growth conditions employed during epitaxy. For example, silicon is not chemically stable and forms rough, intermixed interfaces with BP when the latter is deposited at temperatures above 950 °C.13 Furthermore, high Si concentrations (∼1019 cm−3 or more) incorporate into BP films due to Si diffusion and autodoping from the Si substrate, thereby altering the electrical properties of BP.10,14−17 Although higher temperatures improve the overall quality of BP,9,22 substrates such as GaN/sapphire and Al0.2Ga0.8N/sapphire limit the deposition temperature due to poor thermal stability of GaN and AlGaN films above 1000 °C. BP films grown directly on sapphire above 1000 °C peel-off from the surface19 due to poor adhesion which was witnessed by the present author as well. The large mismatch in lattice constants and thermal expansion coefficients between BP and substrates such as Si and sapphire result in a high density of dislocations and other defects. The high density of dislocations incorporating into the films further degrades the charge transport properties of BP. BP films deposited on Si either crack13 or bow after the Si is removed24,25 due to tensile stresses generated in the films. Received: October 28, 2015 Revised: December 4, 2015

A

DOI: 10.1021/acs.cgd.5b01525 Cryst. Growth Des. XXXX, XXX, XXX−XXX

Crystal Growth & Design

Article

Figure 1. SEM micrographs of BP grown on on-axis AlN/sapphire, depicting a change in grain size and orientation of BP(111) crystallites at different temperatures (a) 1000 °C, (b) 1100 °C, and (c) 1200 °C. Magnification at 10000×. SiC(0001).22 Ultrahigh purity phosphine (99.9999%) and diborane (1% in H2) gases were the phosphorus and boron sources. Hydrogen gas purified through a palladium membrane was used as the carrier gas. BP films were grown on two types of AlN substrates: (1) AlN buffer layers (∼2 μm) deposited on on-axis sapphire(0001) substrate and (2) AlN buffer layers (∼2 μm) deposited on off-axis sapphire(0001) substrate: tilted 0.2° toward the [11̅00] direction; 1° toward the [1120̅ ] direction; and 2° toward the [1120̅ ] direction. Substrates were heated at 1200 °C under H2 flow in order to clean the surface prior to BP deposition. After in situ H2 etching, reactant gases were introduced into the carrier gas stream to deposit the BP films. The growth temperature was varied between 1000 and 1200 °C, and gas flow rates were varied between 4000 and 6000 sccm of H2, 20−100 sccm of PH3, and 30−80 sccm of B2H6 (1% in H2). Films were deposited for 30 min to 3 h. Film morphology was characterized by scanning electron microscopy (SEM) using FEI Nova NanoSEM 230 microscope equipped with field emission gun operated at 15 kV with vCD detector. Crystalline orientation and quality of films deposited at different temperatures and reactant flow rate ratios were determined by X-ray diffraction (XRD). Relative peak intensities of different BP orientations and peak widths of preferred BP(111) orientation were

To date, there are no reports of BP epitaxy on either bulk AlN or AlN/sapphire substrates in the literature. Therefore, to address the issues noted above, AlN(0001)/sapphire was investigated as a substrate for growing BP epitaxial films in this work. AlN has a lower lattice mismatch (3.2%) with BP compared to Si (16.4%), SiC (4.2%), and sapphire (32.6%) substrates. The excellent chemical stability of AlN enables BP epitaxy at elevated temperatures (>1100 °C), potentially improving the film quality. Furthermore, AlN is composed of group III and V elements which are isoelectronic with BP, making unintentional doping less likely. Because of these reasons, AlN has the potential to improve structural and electrical properties of BP. In this paper, the process parameters that affect crystal quality, grain size, film orientation, and electrical properties of BP are reported, analyzed, and discussed.

2. EXPERIMENTAL DETAILS BP films were grown in a horizontal cold-wall chemical vapor deposition reactor at an operating pressure of 700 Torr. The reactor setup was similar to our previous work on BP film growth on 4HB



Article

analyzed by θ/2θ scan using Rigaku Miniflex II X-ray diffractometer with CuKα1 (1.54A) source. Double axis rocking curves (ω scan and ω-2θ scan) were recorded by high resolution X-ray diffraction (HRXRD) using Bede D1 diffractometer with CuKα1 (1.54A) source and beam conditioning using Max-Flux optics and Ge (111) channelcut monochromator to assess the rocking curve widths and mismatch between BP layer and the substrate. The epitaxial relationship, internal stresses, and the presence of rotational twins in BP films were determined by X-ray topography, carried out at beamline X19C at the National Synchrotron Light Source, Brookhaven National Laboratory. Transmission topographs were recorded using white beam radiation that was aligned parallel to the [0001] direction of the AlN(0001)/ sapphire substrate. Raman spectra of the films were recorded using a 0.55 m spectrometer (iHR550, Horiba-Jobin Yvon) that was coupled to an upright microscope (BX41, Olympus) and a light source of 633 nm HeNe laser. The electrical properties of BP between room temperature and 700 K were measured by the Van der Pauw method using the LakeShore Hall measurement system model no. 7704A. Ohmic contacts were made by evaporating the Al metal and annealing in N2 at 400 °C for 1 h.

single crystalline films that were uniform and continuous over the entire surface. Because PH3 was in excess, growth rate was only dependent on the B2H6 flow rate. At a PH3/B2H6 flow rate ratio of 100, films were highly textured with triangular facets but were rough due to the high deposition rate. When the PH3/ B2H6 flow rate ratio was increased to 150 and 200 (by decreasing B2H6 flow rate), the surface roughness of films decreased. Similarly, at 1000 and 1100 °C, the films became smooth when the PH3/B2H6 flow rate ratio was increased from 100 to 200. However, increasing the PH3/B2H6 flow rate ratio did not change the texture and crystalline orientation of the films (see Supporting Information, Figure S2). In summary, at a constant temperature, variation of the PH3 flow rate influenced the crystallinity and continuity of the film on the substrate surface, whereas alteration of the B2H6 flow rate (PH3/B2H6 ratio) influenced roughness and growth rate. Temperature significantly affected the grain size and orientation of the preferential BP(111) crystallites, whereas reactant flow rates affected surface roughness and uniformity of the films. Representative XRD patterns extracted from θ/2θ scans for films deposited at 1000, 1100, and 1200 °C on on-axis AlN/ sapphire substrate are shown in Figure 2. Diffraction peaks

3. RESULTS AND DISCUSSION The grain size and crystalline orientation of BP grown on various types of AlN(0001)/sapphire substrates were evaluated by SEM. For all substrates, grain size and film texture were strongly dependent on temperature and reactants’ flow rate ratio but independent of tilt angle and tilt direction of the sapphire substrate. SEM indicated that crystalline orientation improved with temperatures between 1000 and 1200 °C. SEM micrographs of BP films grown at 1000, 1100, and 1200 °C on AlN/sapphire on-axis substrate are shown in Figure 1. Films grown at 1000 °C or less were composed of a polycrystalline mixture of fine triangular- and irregular-shaped grains with some degree of common alignment. At temperatures above 1000 °C, larger grains were formed with enhanced degree of faceting. At the highest deposition temperature studied (1200 °C), grains were preferentially deposited as large triangles with the base parallel to the substrate surface, indicating highly textured films with an epitaxial relationship between the BP(111) and c-plane of the AlN/sapphire substrate. Grain size measured at 1200 °C (∼15 μm) was approximately 6 times the size measured at 1000 °C (∼2.5 μm). A similar increase in grain size with temperature was seen on other AlN(0001)/ sapphire substrates that were misoriented in different tilt angles and directions (see Supporting Information, Figure S1). Also, it is worth noting that none of the BP films deposited on AlN/ sapphire substrates at different temperatures and reactant flow rate ratios were peeled-off from the substrate. In addition to the temperature, reactants’ flow rates were also shown to be important process variables for improving the crystalline quality of films. The influence of reactant flow rates on surface morphology and roughness was studied by varying the PH3 and B2H6 flow rates separately at a fixed temperature. The minimum PH3 flow rates to prevent BP decomposition were determined to be 30, 40, and 80 sccm at 1000 °C, 1100 °C, and 1200 °C, respectively. Flow rates below these values resulted in amorphous or polycrystalline films with scattered island growth on the surface. In addition, the PH3 flow rate must be kept at least several times (>75) higher than the B2H6 flow rate in order to grow single crystalline films. This was accomplished by maintaining the minimum flow rate of PH3 at the respective temperature, and then B2H6 (1% in H2) flow rate was varied over a range of 20−80 sccm. For example, at 1200 °C, PH3 was maintained at 80 sccm, and the B2H6 (1% in H2) flow rate was varied between 40 and 80 sccm in order to obtain

Figure 2. XRD patterns (θ/2θ scans) showing the effect of temperature on crystalline orientation of BP. Dash lines and dotted lines indicate peaks from preferred BP orientation and AlN(0001)/ sapphire substrate, respectively.

from the substrate occur at 36.04°, 76.44°, and 41.68° in the figure, representing the AlN(0002), AlN(0004), and sapphire(0006) planes, respectively. Intense peaks at 34.30° and 72.03° at all temperatures show the preferential BP(111) and BP(222) orientations, respectively. Additional peaks at 39.67°, 57.4°, and 68.54°, corresponding to BP(200), BP(220), and BP(311) orientations, respectively, along with few minor unknown peaks below 34° were also visible for films deposited at 1000 °C, indicating that these films had some degree of C



Article

In order to corroborate the above findings, HRXRD rocking curves (ω-scan) and ω-2θ scans were performed on BP films deposited on both on-axis and off-axis AlN/sapphire substrates. As shown in Figure 4, the FWHM values of BP(111) ω rocking

polycrystallinity. However, the intensities of these additional peaks were significantly smaller than the preferred BP(111) orientation. At 1100 and 1200 °C, BP(111) orientation was predominant, and additional BP orientations were negligible, thereby validating SEM results that the films were single crystalline. The full width at half-maximum (FWHM) values of preferred BP(111) peak and relative peak intensity ratios of BP(111) and other BP orientations extracted from XRD θ/2θ scans on films deposited on AlN/sapphire tilted 1° toward the [112̅0] direction at different reaction conditions are shown in Figure 3. Peak widths of BP(111) decreased as growth

Figure 4. BP(111) ω rocking curves measured on (a) AlN/sapphire tilted 1° toward the [1120̅ ] direction and (b) AlN/sapphire on-axis substrates.

curves consistently decreased with an increase in growth temperature and PH3/B2H6 ratio, indicating the best crystal quality obtained at 1200 °C and a PH3/B2H6 flow rate ratio of 200. Furthermore, a similar trend in decreasing peak widths with temperature and PH3/B2H6 flow rate ratio for the same samples was observed from the ω-2θ scans and Raman spectroscopy (Table 1). Residual strain in these films was assessed by measuring the lattice mismatch between BP(111) and AlN(0002) planes from the ω-2θ scans. After comparing the theoretical lattice mismatch with the measured lattice mismatch values as shown in Table 1, it can be concluded that BP films grown on AlN/sapphire substrates were mostly relaxed, although some residual compressive strain may still be present. Raman spectra of BP films grown on AlN/sapphire tilted 0.2° toward the [11̅00] direction at 1000 °C, 1100 °C, and 1200 °C for 30 min are shown in Figure 5. Raman peaks characteristic of TO and LO phonon modes of BP were evident at 802 and 832.6 cm−1, respectively. Raman wavenumbers measured for BP were in good agreement with values reported in the literature.14,22,26,27 The intensity of LO mode is typically stronger than the TO mode, so peak widths were calculated only for the LO mode. FWHM values consistently decreased with increase in deposition temperature, and peak width (FWHM = 6.1 cm−1) measured at 1200 °C on this substrate was lower than values previously reported in the literature for BP epitaxial films.22 Similarly, peak widths decreased for a majority of the samples when the PH3/B2H6 flow rate ratio increased (Table 2). The magnitude of decrease in FWHM values was more pronounced with increased temperature as compared to the increased P/B reactant flow ratio. A similar trend in reduced peak widths with temperature and reactant flow ratios was also observed for BP films grown on 4HSiC(0001) under identical conditions in our previous work.22 However, the FWHM values calculated at identical temperatures and reactant flow rates were consistently narrower for AlN/sapphire compared to 4H-SiC(0001) substrate. The decrease in FWHM values with temperature on other types of AlN/sapphire substrates are shown in Supporting Information (Figure S3). Raman peaks shifted slightly to higher wavenumbers with increasing temperature, possibly due

Figure 3. (a) FWHM values of BP(111) peak and relative peak intensities of (b) BP(111)/BP(200), and (c) BP(111)/BP(220) extracted from XRD θ/2θ scans.

temperature was increased from 1000 to 1200 °C at a constant reactant flow rate ratio (Figure 3a). This decrease in FWHM is perhaps due to the increase in crystallite size with temperature since FWHM values are inversely proportional to crystallite size. Overall results showed that the FWHM decreased with temperature and PH3/B2H6 ratio; the lowest value of 0.18° occurred at 1200 °C and PH3/B2H6 = 200. Similarly, the peak intensity ratios of BP(111)/BP(200) and BP(111)/BP(220) increased with temperature and PH3/B2H6 flow rate ratio due to the improved texture and increased BP(111) peak intensity, as shown in Figure 3b,c. At low temperatures, adatoms on the surface will be less mobile and hence deposit as individual fine grains during the initial growth phase. These fine individual grains with large numbers of grain boundaries act as multiple nucleation sites and enable the BP crystallites to grow in other crystallographic planes, such as (200), (220), and (311). Hence, polycrystalline films were obtained at low temperatures in this study. In contrast, high temperatures (above 1000 °C) enable more movement of the adatoms, resulting in the expansion and coalescence of small individual grains to form large grains with preferential BP(111) orientation. Similarly, when the diborane flow rate was reduced, surface smoothness was enhanced due to the decreased deposition rate. Hence, highest θ/2θ peak ratios of BP(111)/BP(200) = 5157 and BP(111)/BP(220) = 7226 occurred at a high temperature (1200 °C) and high PH3/B2H6 ratio of 200. D



Article

Table 1. BP Peak Widths and Strain Evaluation of Films Grown at Various Reaction Conditions substrate AlN/sapphire, tilted 1° toward [112̅0] AlN/sapphire, on-axis

temp (°C)

PH3/ B2H6

1000 1100 1200

200

1200

100 200

ω-2θ scan FWHM (arcsec)

Raman FWHM (cm−1)

551 390 249 381 303

10.8 7.9 6.8 7.2 7.1

theoretical mismatch (%)

mismatch from ω-2θ scan (%)

5.22

5.01 5.18 5.30 5.25 5.18

Figure 6. X-ray transmission topograph of BP grown on on-axis AlN(0001)/sapphire substrate. Diffraction spots shown by circles, hexagons, diamonds, and squares indicate BP(111), twinned BP(111), sapphire(0001), and AlN(0001) planes, respectively.

toward [112̅0] direction. BP films measured at room temperature had n-type conductivity with a carrier concentration in the range of 1018 cm−3 (Table 3). The electron

Figure 5. Raman spectra showing optical modes of BP and FWHM values of LO mode at different temperatures.

Table 2. FWHM of Raman Peaks for BP Grown on AlN/ Sapphire Tilted 1° toward [112̅0] Direction at Different Deposition Conditions

Table 3. Electrical Properties of BP Films Grown at 1200 °C Measured at Room Temperature

FWHM (cm−1) temp (°C)

PH3/B2H6 = 100

PH3/B2H6 = 150

PH3/B2H6 = 200

1000 1100 1200

11.2 7.9 7.3

10.8 8.4 7.0

10.1 7.3 6.8

sample No. No. No. No. No.

to increased residual strain generated by cooling the BP films over a wider temperature range. BP films were analyzed by synchrotron white beam X-ray topography to assess the epitaxial relationship and formation of BP(111) twin orientations. Figure 6 shows the topograph recorded on BP film (∼3 μm) grown on on-axis AlN(0001)/ sapphire substrate. The diffraction spots of BP(111), AlN(0001), and sapphire(0001) confirm the existence of an epitaxial relationship between BP and the substrate. The epitaxial relationship between BP and AlN was (111)BP ̅ BP || (0001)AlN AlN. The spots indicated by hexagons confirm the presence of BP(111) rotational twins, the most commonly observed defects among the III−V compound semiconductors.9,20,22,28 These defects were formed due to the symmetry mismatch between BP(111) with 3-fold symmetry and underlying AlN(0001) substrate with 6-fold symmetry. Preliminary Hall effect measurements were performed on undoped BP films grown at 1200 °C on AlN/sapphire tilted 1°

1 2 3 4 5

PH3/B2H6 100 150 200 200 200

growth time (min) 30 30 30 30 180

type

n (cm−3)

μ (cm2/V·s)

n n n n n

× × × × ×

14.2 15.4 18.0 35.3 27.5

8.76 6.04 5.25 4.50 3.39

18

10 1018 1018 1018 1018

mobility slightly increased as PH3/B2H6 ratio was increased from 100 to 150 (No. 1−3). For comparison, BP films were deposited on a polished AlN layer (No. 4 and 5) under identical conditions to evaluate the influence of surface roughness of the starting substrate. Mobility increased significantly, perhaps due to the deposition of BP on a smoother AlN layer. Hall measurements were performed on a thicker film (∼12 μm) grown on a polished AlN layer (No. 5) in order to assess the effect of film thickness on mobility. Mobility decreased slightly from 35.3 cm2/V·s to 27.5 cm2/V·s, but was still higher than the mobility of films grown on unpolished AlN substrates (Nos. 1−3). Since the 12 μm thick film was crack-free and had reasonably high mobility, film thickness could be further increased to test BP performance in neutron detection applications. E



■

The temperature dependence of electron concentration and mobility are shown in Figure 7a,b. The intrinsic electron

Article

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.cgd.5b01525. Scanning electron micrographs; Raman spectra (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Address: Department of Chemical Engineering Kansas State University 1005 Durland Hall, 1701A Platt St Manhattan, KS, USA 66506. Phone: 785-532-4320. Fax: 785-532-7372. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors would like to thank the U.S. Department of Energy for funding this research under Grant No. DESC000516. X-ray topography was carried out at Stony Brook Topography Facility (Beamline X19C) at the NSLS, Brookhaven National Laboratory, which is supported by the U.S. DOE under Grant No. DE-AC02-76CH00016. We are grateful to Nitride Solutions Inc., Wichita, Kansas, for generously providing the AlN/sapphire substrates for this research and to Dr. Placidus Amama and Nolan Gaede for helping us with metal contacts deposition for Hall measurements. Raman spectroscopy was supported by the National Science Foundation under grant no. EPSCoR IIA−1430493.

Figure 7. Temperature dependence of (a) carrier concentration and (b) Hall mobility of BP films deposited at 1200 °C on AlN/sapphire tilted 1° toward [112̅0] direction.

concentration increased exponentially with increasing temperature. The solid lines represent the least-squares fit to the experimental data per the relation, n (T) ∝ exp(−ED/kT), where n (T) is electron concentration as a function of temperature. The unintentional donor activation energy (ED) calculated from the slope of ln(n) vs (1/T) plot was in the range of 0.02−0.03 eV. Hall mobility of electrons increased with temperature initially, reaching the maximum value at approximately 450 K, and then decreased slowly with further increase in temperature. The highest mobility was consistently above the room temperature values for all samples. The mobility trend in the entire temperature range was difficult to interpret because neither empirical laws μ ≈ T−3/2 nor μ ≈ (T3/2/NI) satisfactorily explain the observed phenomenon. These preliminary mobility measurements indicate that more than one scattering mechanism was taking place. More detailed measurements needs to be performed in order to understand this phenomenon in detail.

■

REFERENCES

(1) Ejembi, J. I.; Nwigboji, I. H.; Franklin, L.; Malozovsky, Y.; Zhao, G. L.; Bagayoko, D. J. Appl. Phys. 2014, 116 (10), 103711. (2) Stone, B.; Hill, D. Phys. Rev. Lett. 1960, 4 (6), 282−284. (3) Takenaka, T.; Takigawa, M.; Shohno, K. Jpn. J. Appl. Phys. 1976, 15 (11), 2235−2236. (4) Kumashiro, Y.; Mitsuhashi, T.; Okaya, S.; Muta, F.; Koshiro, T.; Takahashi, Y.; Mirabayashi, M. J. Appl. Phys. 1989, 65 (5), 2147−2148. (5) Wentzcovitch, R. M.; Chang, K. J.; Cohen, M. L. Phys. Rev. B: Condens. Matter Mater. Phys. 1986, 34 (2), 1071−1079. (6) Bouhafs, B.; Aourag, H.; Ferhat, M.; Certier, M. J. Phys.: Condens. Matter 1999, 11 (30), 5781−5796. (7) Archer, R. J.; Koyama, R. Y.; Loebner, E. E.; Lucas, R. C. Phys. Rev. Lett. 1964, 12 (19), 11−13. (8) Wang, C. C.; Cardona, M.; Fischer, A. G. RCA Rev. 1964, 25, 159−167. (9) Chu, T. L.; Jackson, J. M.; Hyslop, A. E.; Chu, S. C. J. Appl. Phys. 1971, 42 (1), 420−424. (10) Kumashiro, Y. J. Mater. Res. 1990, 5 (12), 2933−2947. (11) Shohno, K.; Takigawa, M.; Nakada, T. J. Cryst. Growth 1974, 24−25, 193−196. (12) McGregor, D. S.; Unruh, T. C.; McNeil, W. J. Nucl. Instrum. Methods Phys. Res., Sect. A 2008, 591 (3), 530−533. (13) Schroten, E.; Goossens, A.; Schoonman, J. J. Appl. Phys. 1998, 83 (3), 1660. (14) Kumashiro, Y.; Okada, Y.; Okumura, H. J. Cryst. Growth 1993, 132 (3−4), 611−613. (15) Ohsawa, J.; Nishinaga, T.; Uchiyama, S. Jpn. J. Appl. Phys. 1978, 17 (9), 1579−1586. (16) Takenaka, T.; Takigawa, M.; Shohno, K. J. Electrochem. Soc. 1978, 125 (4), 633. (17) Kim, C. J.; Shono, K. J. Electrochem. Soc. 1984, 131 (1), 120− 122. (18) Kumashiro, Y.; Nakamura, K.; Enomoto, T.; Tanaka, M. J. Mater. Sci.: Mater. Electron. 2011, 22 (8), 966−973.

4. CONCLUSIONS This study investigated the growth of epitaxial BP films on AlN(0001)/sapphire substrates with on- and off-axis orientations. We successfully demonstrated the growth of high quality single crystalline BP films on AlN(0001) substrate over a temperature range of 1000−1200 °C. The crystalline quality of the films was dependent on temperature and PH3/B2H6 flow rate ratio, which influenced the crystalline orientation, grain size, growth rate, and surface roughness of the films. High deposition temperatures increased the BP(111) crystallites’ grain size and orientation, while high PH3/B2H6 flow rate ratios controlled the growth rate, resulting in smoother films with enhanced crystal quality. The BP films had rotational twins due to mismatch of crystal symmetry between BP(111) and AlN(0001) planes. The deposited films were almost completely relaxed. The films adhered well to the substrates and were crack-free. Narrow BP peak widths from XRD and Raman spectroscopy, as well as large Hall mobilities at high carrier concentrations, confirmed that AlN is a superior substrate for epitaxial BP films with promising potential for further enhancement of BP properties. F



Article

(19) Kumashiro, Y.; Matsumoto, M.; Yoshizawa, H. JJAP Ser. (1994), Proc. 11th Int. Symp. Boron, Borides Relat. Compd. 1994, 10, 160−163. (20) Udagawa, T.; Odawara, M.; Shimaoka, G. Appl. Surf. Sci. 2005, 244 (1−4), 285−288. (21) Li, G.; Abbott, J. K. C.; Brasfield, J. D.; Liu, P.; Dale, A.; Duscher, G.; Rack, P. D.; Feigerle, C. S. Appl. Surf. Sci. 2015, 327, 7− 12. (22) Padavala, B.; Frye, C. D.; Ding, Z.; Chen, R.; Dudley, M.; Raghothamachar, B.; Khan, N.; Edgar, J. H. Solid State Sci. 2015, 47, 55−60. (23) Ganesan, S.; Maradudin, a.; Oitmaa, J. Ann. Phys. 1970, 56, 556−594. (24) Suzuki, A.; Takigawa, M.; Shohno, K. Jpn. J. Appl. Phys. 1977, 16 (6), 1053−1054. (25) Nishinaga, T.; Ogawa, H.; Watanabe, H.; Arizumi, T. J. Cryst. Growth 1972, 13−14, 346−349. (26) Schmitt, J. O.; Edgar, L. J. H.; Liu, L.; Nagarajan, R.; Szyszko, T.; Podsiadlo, S.; Wojciech, G. Phys. Status Solidi C 2005, 2 (3), 1077− 1080. (27) Sanjurjo, J.; Lopez-Cruz, E.; Vogl, P.; Cardona, M. Phys. Rev. B: Condens. Matter Mater. Phys. 1983, 28, 4579−4584. (28) Zhang, Y.; Chen, H.; Dudley, M.; Edgar, J. H.; Gong, Y.; Bakalova, S.; Kuball, M.; Zhang, L.; Su, D.; Zhu, Y. J. Cryst. Growth 2012, 352 (1), 3−8.

G


Sapphire

Recommend Documents